From: Uwe Freese <mail_at_uwe-freese.de>
Date: Tue, 10 Dec 2002 18:12:24 +0100

Hello Josh,

Josh (Norm) Audette wrote on Tuesday, December 10, 2002, 3:02:06 PM:

JNA> The common definition of trickle charging is to allow a small current
JNA> to continuously charge the battery. With larger currents, rechargeable
JNA> You'd have to read the specs for the battery type to determine what a
JNA> proper trickle charge current is (I believe the links in the charging

Usual "trickle charge" (used in other battery chargers) means to charge
a small amount that is exactle the self-discharging amount of the
batteries.
Charging to prevent the self-decharging is usually ~1/20 C (with C =
capacity/h).

The self-discharge cycle goes over weeks (NiMH) or even months (NiCd,
Alcaline batteries). It is not so important how big the charging
current (somewhere between 1/20 C to 1/100 C) is.

What we want for rockbox is to charge with the amount an archos jukebox
usually consumes (that is ~100mA). (The reason for charging is because
the archos is on and not the self-discharging of the batteries.)
That means a shorter time till the batteries are would be empty and a
bigger charging current to avoid this. I think my idea is good because
it adjusts the charging current dynamically and I think here it is more
important that the charge current fits the decharge current "exactly".

Another word to detection of full batteries, as you told usually done
(and also in rockbox) with dV/dt. This is useful for a normal charge
cycle but not for a trickle charge. The current is too small and I
don't want the batteries' voltage go so high that one can detect a "delta
peak" every few hours! That's not good for them.

So I think
- complete charge cycle
- trickle charge (as I described in the old mail, with dynamically small
  current) max. 10h or so
- then decharge
- start charge cycle again if voltage falls under the
  start-charge-voltage
is best.
  
Bye, Uwe.
Received on 2002-12-10